Elven Hair and Eye Color

Faelivrin

Well-Known Member
The chances for each eumelanin gamete from each genome look like this:


10 VVWWXXYYZZ = deep black 5 (100%)
9 VVWWXXYYZz = “black” 4 (50%), 5 (50%)
8 VVWWXXYyZz = dark brown with rr = chestnut 3 (25%), 4 (50%), 5 (25%)
7 VVWWXxYyZz = brown with rr = dark auburn 2 (12.5%), 3 (37.5%), 4 (37.5%), 5 (12.5%)
6 VVWwXxYyZz = brown with rr = light auburn 1 (6.25%), 2 (25%), 3 (37.5%), 4 (25%), 5 (6.25%)
5 VvWwXxYyZz = light/golden brown with rr = copper 0 (3.125%), 1 (15.625%), 2 (31.25%), 3 (31.25%), 4 (15.625%), 5 (3.125%)
4 VvWwXxYyzz = sandy/ash blond with rr = ginger 0 (6.25%), 1 (25%), 2 (37.5%), 3 (25%), 4 (6.25%)
3 VvWwXxyyzz = deep gold with rr = ginger 0 (12.5%), 1 (37.5%), 2 (37.5%), 3 (12.5%)
2 VvWwxxyyzz = golden blond with rr = ginger 0 (25%), 1 (50%), 2 (25%)
1 Vvwwxxyyzz = light blond/flaxen with rr = strawberry blond 0 (50%), 1 (50%)
0 vvwwxxyyzz = platinum blond with rr = strawberry blond 0 (100%)

In theory, I can multiply the %s, /100, of the 2 gametes for any given combo, then add up the %s of combos that produce the same result, to get the chances from any given result.


Every possible cross for Feanor + Nerdanel, with the gamete %s: (I wish I knew how to insert a table)

Fëanor 9 × Nerdanel 5: 10 (1.6%), 9 (9.4%), 8 (23.4%), 7 (31.3%), 6 (23.4%), 5 (9.4%), 4 (1.6%)
Fëanor 9 × Nerdanel 6: 10 (3.1%), 9 (15.6%), 8 (31.3%), 7 (31.3%), 6 (15.6%), 5 (3.1%) (reproduced/confirmed your result)
Fëanor 10 × Nerdanel 5: 10 (3.1%), 9 (15.6%), 8 (31.3%), 7 (31.3%), 6 (15.6%), 5 (3.1%)
Fëanor 9 × Nerdanel 7: 10 (6.3%), 9 (25%), 8 (37.5%), 7 (25%), 6 (6.3%)
Fëanor 10 × Nerdanel 6: 10 (6.3%), 9 (25%), 8 (37.5%), 7 (25%), 6 (6.3%)
Fëanor 10 × Nerdanel 7: 10 (12.5%), 9 (37.5%), 8 (37.5%), 7 (12.5%)

Nerdanel: +r, +?, 5-7 brown
Fëanor: ?r, +s, 9-10 black
Maedhros: rr, +?, 5-7 or rr, +s (diluted) 6-8 copper or auburn
Maglor: ??, ??, 5-10 dark
Celegorm: ??, ??, 5-7 brown
Caranthir: ??, +?, 6-10 dark
Curufin: ??, +?, 6-10 dark
twins: rr, +s (diluted) 7-8 copper/auburn/dark

I like the chances from 9x6 and 10x5 best. Very low chances of a 10 or 5, but still Celegorm could be 5, Maedhros 6, and Maglor-Caranthir-Curufin 7-9.
Otherwise, 9x7 or 10x6, with Maedhros and Celegorm diluted somehow via carrying silver.
9x5 makes 5 (copper or light/golden brown) more likely and 9 equally so, with 4 and 10 as very unlikely outliers.



RE: Indis x Finwe, I like giving Indis a 3. In think somewhere Fingolfin is called "his father's son, tall, dark, and proud". I wouldn't want him lighter than an 8 or 7. Irime can be a lighter brown (5-7). Also I imagine that, among Elves, 4 is a deep gold instead of the greyish ash-blond Humans get. And I'm going to assume Indis' parents and Elenwe are no darker than 4.

That still leaves Fingolfin's kids wide open because we know nothing about Anaire, except she's Noldorin. I'd prefer her and Turgon and Eol to be dark, but they aren't specified AFAIK. I'm going to cross Fingolfin (7-8) x Anaire (5-10), and then Turgon x Elenwe (0-4), and Aredhel x Eol. (This is so weird. I feel like I'm talking about breeding rabbits. Actually doing so with rabbits named after Elven princes would be... a level of fannish dedication I'm not prepared to go to.)
 
Last edited:

MithLuin

Administrator
Staff member
(I wish I knew how to insert a table)
What I am doing is making a table in Excel, highlighting it with the cursor, selecting the 'copy as picture' setting, opening MS Paint, pasting, crop to size, save as a jpeg, and then uploading the file here.

Yeah, it's not easy - so convoluted!



(This is so weird. I feel like I'm talking about breeding rabbits. Actually doing so with rabbits named after Elven princes would be... a level of fannish dedication I'm not prepared to go to.)
I am not starting a rabbit breeding program! :p Yeah, it's weird to talk about humans using language designed for farm animals, so I either try to stick with technical, or slip in the occasional marriage rather than cross.
 

Faelivrin

Well-Known Member
Where's Celebrimbor called "dark"?

And do you think it's likely Celebrian and/or Nimloth were silver-haired?



Turgon x Elenwe now in progress...
Probably not going to take the effort to insert pics tho.
 
Last edited:

MithLuin

Administrator
Staff member
I don't know that Tolkien ever described Celebrimbor (though if he did, it would have likely been to emphasize his similarity to Fëanor). I was mostly commenting that in his case, the 'celeb' in his name definitely does not refer to his hair color. It means 'silver fist'. So, dark hair.
 

Faelivrin

Well-Known Member
Considering that Celebrimbor has also been called an elf of Gondolin, a Sinda descended from Daeron, and a Teler from Alqualonde, any description of him not explicitly connected to the House of Feanor should be... suspect.

Anyway I finished Fingolfin x Anaire but it isn't very useful until I get some constraints on what Turgon and Aredhel can be.


Fingolfin: ??, +?, 7-8 dark brown
Anairë: ??, ??, 5-10 dark?

8x10 = 10 (25%), 9 (50%), 8 (25%)
7x10 = 10 (12.5%), 9 (37.5%), 8 (37.5%), 7 (12.5%)
8 x 9 = 10 (12.5%), 9 (37.5%), 8 (37.5%), 7 (12.5%)
7 x 9 = 10 (6.3%), 9 (25%), 8 (37.5%), 7 (25%), 6 (6.3%)
8 x 8 = 10 (6.3%), 9 (25%), 8 (37.5%), 7 (25%), 6 (6.3%)
7 x 8 = 10 (3.1%), 9 (15.6%), 8 (30.9%), 7 (30.9%), 6 (15.6%), 5 (3.1%)
8 x 7 = 10 (3.1%), 9 (15.6%), 8 (30.9%), 7 (30.9%), 6 (15.6%), 5 (3.1%)
7 x 7 = 10 (1.6%), 9 (9.4%), 8 (23.4%), 7 (31.3%), 6 (23.4%), 5 (9.4%), 4 (1.6%)
8 x 6 = 10 (1.6%), 9 (9.4%), 8 (23.4%), 7 (31.3%), 6 (23.4%), 5 (9.4%), 4 (1.6%)
7 x 6 = 10 (0.8%), 9 (5.5%), 8 (16.4%), 7 (27.3%), 6 (27.3%), 5 (16.4%), 4 (5.5%), 3 (0.8%)
8 x 5 = 10 (0.8%), 9 (5.5%), 8 (16.4%), 7 (27.3%), 6 (27.3%), 5 (16.4%), 4 (5.5%), 3 (0.8%)
7 x 5 = 10 (0.4%), 9 (3.1%), 8 (10.9%), 7 (21.9%), 6 (27.3%), 5 (21.9%), 4 (10.9%), 3 (3.1%), 2 (0.4%)
 
Last edited:

MithLuin

Administrator
Staff member
Conclusions I've reached thus far:

Finwë is a 9.
Fëanor is a 9.
Nerdanel is a 5-6.
Maedhros is a 5.
Curufin is a 9.
Indis is a 3.
Fingolfin is an 8.
Finarfin is a 4.
Findis is a 5.
Irime is a 7.

Guess it's time to tackle silver next....
 

MithLuin

Administrator
Staff member
When I wrote Nimloth, I gave her pale hair, and compared her to a moth. (I had to go re-read that passage in my own story to see what choice I made ;) ). So, yes, my first inclination would be to make Nimloth silver-haired. She is related to several elves who are known to have silver hair, and she has a name with a similar structure to Celeborn. So...yes, that seems likely. It does not seem to be 100% necessary, but it's a good first guess.

Same with Celebrían. With her parents having silver hair and gold hair with silver highlights, it seems extremely likely that the 'celeb' in her name refers to her hair color.

So, I'd certainly like to pick an inheritance pattern for silver that allows these characters to have silver hair, though there are of course other options for them.
 

Faelivrin

Well-Known Member
Finarfin's kids aren't very constrained... Angrod needs to be light enough to have a blond kid and at least one blond grandkid. Galadriel has dark grandkids, but I'm increasingly inclined to credit Melian with simple dominant super-black hair.

Agreed Fingolfin is probably 8, but wait till I get to his grandchildren (unless you already have).

Still working on Turgon x Elenwe and planning to do Aredhel x Eol, but I'm now tired of calculating all these probabilities and want to program giant Punnett squares into Excel before I continue. This'll take a while, but then I will have the % chances of every outcome of every cross -- although based on my assumption of maximum heterozygosity in every genome, so it won't be as accurate/detailed as any of your Punnett squares. But if you don't mind that fudge/shortcut, it'll save a lot of time.
 
Last edited:

MithLuin

Administrator
Staff member
So far, we are accounting for all shades of black, brown and blond with our 5 incomplete dominant additive pigmentation genes(VWXYZ), giving a range of 11 different possible shades, which can be matched to natural shades of human hair color.

In addition, we are using a recessive red gene (r) which can override blond and light shades of brown, but is masked by darker hair.

For silver, I want to consider treating it as a gene which negates pigment, washing it out. Dilution genes are quite common in many animals. For instance, the difference between a black cat and grey cat, or a bright orange and a pale orange.





An alternative is that silver is the result of a unique silver gene, or that silver is an alternative allele in the pigmentation that is not the typical dominant/recessive, but has another effect. That is trickier to work out, because you have to combine it with the VWXYZ genes. Since there's already 5 of those to keep track of, and they're annoying enough to work with, I am going to leave out the 'alternate allele' option for now.


In cats, the dilution gene affects all colors. So, black becomes 'blue' (grey), orange becomes cream, and calico becomes dilute calico. In other words, it would affect all ranges, 0-10 of the black/brown/blond colors, as well as the red. But simply diluting pigmentation doesn't actually give silver hair. So we would have to figure out what type of silver hair we're talking about, and how a dilution gene could achieve it.

It is possible to envision pale silver hair as a pale blond hair that has been bleached out a bit. And so, some of the pigment removed. But there are certainly darker shades of silver that would look more like the grey cats, a 'washed out' black. And pure white hair would not have any pigment at all, so if silver means white, then we could be looking at completely blocking the pigment, rather than simply diluting it, which would be more of an albino hair color.

I think that dilution in horses is more informative than in cats. Horses can have multiple dilution genes with a range of effects. Horses can also have grey or white as a natural coat color (not simply old age), so probably a better basis for our elvish silver anyway.

Quick overview of horse hair color genetics - they're either red (chestnut) or black (EE, Ee = black, ee=red). Then, they have a multitude of other genes that influence the expression of that base coat. So, Agouti makes black --> bay. Cream dilution is an incomplete dominant that makes a black horse smoky black (1 copy) or smoky cream (2 copies), and a bay horse buckskin (1 copy) or perlino (2 copies). That same cream dilution will make a chestnut horse palomino (1 copy) or cremello (2 copies). There are separate dilution genes to create dun, pearl, and champagne coats. The silver dapple dilution gene ONLY affects black, not red.

In horses, the grey coat is a dominant depigmentation gene. So, a horse that inherits this gene would have been bay or chestnut or whatever, but instead has an awful lot of white hair. These horses do get more white as they age, so they are darker when they are younger. Also, they can have spots/patterns in the grey for lots of variation.



So, we can create a separate silver depigmentation gene, S. I would have to play with some crosses to see what the rules would be, and what phenotype we should assign to SS, Ss, and ss.



For our Galadriel conundrum, it will be useful to consider the Roan trait (seen in cows, horses, dogs, and guinea pigs). This is where two colors (white and something else) are evenly mixed throughout the animal's coat. Because Galadriel is the ONLY elf ever described with this two-tone silver-gold hair, the trait would have to be incredibly rare. In horses, roan is a simple dominant, meaning that the parent would also show the trait. In cattle, roan is only present in heterozygous offspring - you breed a white cow to a red bull to get a strawberry roan calf (for instance). Roan is also incompletely dominant in guinea pigs, but the recessive trait (white) is lethal.

Luckily for us, Galadriel has a background that includes Vanyar, Noldor, and Teleri genetics. So, it is possible for her to get a unique combination that no one else gets without impossibly straining credulity. Also, whatever inheritance pattern we choose, the alleles to produce it will have to be incredibly rare. So, she just got lucky and had parents who could pass on that trait to her. Options: Simple dominant is out. Her father or mother would have to show the trait. Incomplete dominant is tricky. If she has to be Rnrn to show the trait, then there is the chance that this trait would get passed on in her bloodline. We should see her hair again later. However, she only has one child, so it's quite easy to make that side work. The real issue is that, again, one of her parents would either also be Rnrn, or would be pure recessive (rnrn). And we might expect to see this trait in some of her other more distant relatives. Simple recessive is also a possibility, even if that's not how roan works in animals. This is easier. Both of her parents could be (incredibly rare) carriers. So, they had a 1/4 chance of producing an offspring with that trait, and, low and behold...Galadriel. She marries someone who does not carry the trait, and it is again masked in her offspring. If Indis and Olwë's wife are the original carriers, and Fingolfin did not get the allele, this all works out. Like so:

(It occurs to me that I should not use 'R' for roan, since we already have 'r' for red. You typically would use 'Rn', but
I dislike having to go to two letters/symbols. So, 'T' for tiger stripes it is :p)

Indis: Tt
Finwë: TT
Fingolfin: TT (50%)
Finarfin: Tt (50%)
Findis: ?
Irime: ?
Olwë: TT
Olwë's wife: Tt
Earwen: Tt (50%)

Now...while it is true that Finarfin and Earwen only had a 25% chance of having a tt child like Galadriel, they also only had a 25% chance of having a TT child like Angrod. They had a 50% chance of having children who are Tt - who would not show the trait, but *could* pass it on to their children. And so....

Finrod: TT (25%) or Tt (50%)
Angrod: TT (25%) or Tt (50%)
Aegnor: TT (25%) or Tt (50%)
Galadriel: tt (25%)

It is 'safe' for us to allow Finrod and Aegnor to be carriers of this trait, because they do not have children. So, the trait is not passed on. The ONLY individual on the family tree who has the trait is Galadriel. And I recognize that we could keep Angrod as a carrier, but just ensure that Orodreth came out as TT to make sure the trait is not passed on, but you get the idea. And really, they could all be carriers...and it wouldn't matter, because the chance of them marrying another carrier is incredibly low (unless they marry a relative).

Celeborn: TT (Notice I made Olwë's wife the source of the original allele; not Olwë...this doesn't run in Elwë's family)
Celebrian: Tt
Elrond: TT
Elrond's children have a 50/50 chance of being carriers, but none of them would express the trait.

Almost every elf out there is TT, so typically we don't even need to account for this. Galadriel just happened to have two parents carrying a rare recessive allele.


Is it cheating to invent an entire trait and gene just for Galadriel? Yes, of course it is. But since she's the only one with that trait, it's the easiest way to account for it. I realize that Melian's 'super black' is going to create a similar issue. But it's possible that since silver is not part of the black-blond pigmentation scale, we can somehow make that work without it. For instance, Thingol could have a lot of pigmentation genes, but they're all turned off by silver.
 
Last edited:

MithLuin

Administrator
Staff member
Determining an inheritance pattern for silver hair.

First, I need a list of all elves who express the silver hair trait. It should be straightforward to determine if we ever have a silver-haired elf without a silver-haired parent, or if it skips generations to re-appear later in the family tree. 'Should' being the operative word there....

Some known silver-haired elves:
Míriel
Elwë
Celeborn
Círdan
Earwen

'Suspected' silver-haired elves:
Celebrían
Nimloth
Galathil

Olwë explicitly has 'white' hair, not silver. So, we should allow for some variety in our silver trait, with the most washed-out variation being white. Or, he could have 0 out of 10 pigmentation genes, leaving a white coloration...except that should mean his brother would also have the identical pure recessive pigmentation, and he does not. Nimloth also may have this 'white' hair.

Elwing is an interesting question. She is portrayed with a variety of hair colors in art (everything from dark to silver to light brown). Her sons' dark hair definitely came from her (as Tuor and Idril were both blond, necessitating a blond Earendil), but if we count silver as a 'masking' trait, then it would be possible for her to have had significant pigment, but with it completely washed out to silver.

We know nothing about Míriel's family, only her descendants. So, *if* we wanted a silver-haired Celegorm, we would need to make the trait skip a generation to reach him (ruling out simple dominant as an inheritance pattern). Other than that, none of Míriel's descendants are said to have silver hair. I am thinking that a light-brown-haired Celegorm is much more likely.

Círdan is useless, genetically speaking, as he seems to have no relatives at all. He is, presumably, somehow related to Elwë and Olwë, but there is no indication of how closely that might be (he does not seem to be their father or their uncle or anything).

We do see a lot of direct inheritance, not skipping generations.
So, Olwë --> Earwen, Galathil --> Nimloth, and Celeborn --> Celebrían

So perhaps a simple dominant? SS or Ss = Silver, and ss = non-silver This would mean that two dark/blond/red-haired parents could never have a silver-haired child. I don't see that being a problem.

Olwë is SS or Ss. [silver]
His wife is ss.
Their daughter Earwen is Ss. [silver]
Finarfin is ss.
All of Earwen and Finarfin's children are ss. (They had a 50/50 chance)
The silver trait then leaves their family, not showing up in later generations (from this side of the family, anyway).

Elwë is Ss. [silver]
Melian is ss.
Lúthien is ss.
This becomes a problem if 'Dior the Fair' was meant to indicate light hair. But, considering his parentage, it's fair to think it probably meant 'Dior the Beautiful' in this case.
Dior: ss
Nimloth: Ss [silver]
Their children have a 50/50 chance of being Ss [silver] or ss.
Tuor: ss
Elwing's children: ss
Celeborn: Ss [silver]
Galadriel: ss*
Celebrían: Ss [silver]
Elrond's children: ss (they had a 50/50 chance)

*note that I am accounting for Galadriel's gold-and-silver-mixed-strands independently of this trait. As far as the S/s trait goes, Galadriel has blond hair, not silver.

It is easy to think that some of the people in this family tree must be homozygous for the dominant trait. But in reality, silver hair keeps marrying back into the family, so it's difficult to see the trait 'wash out' over generations.


I want to allow for some variations in shade/pigment for the silver.
And so....
If you are 0-3 for pigment, you have 'white' hair if you have the S allele.
If you are 4-7 for pigment, you have 'silver' hair if you have the S allele.
If you are 8-10 for pigment, your dark hair merely dilutes 2 levels of pigment (10 -->8, 9 -->7, 8-->6).
Not sure I like this, but as a starting point, anyway.... (No doubt I'll wind up pulling my hair out trying to get Nimloth and Olwë to be white-haired this way.)


Another possibility is that the R gene for red hair and the S gene for silver are actually variant alleles of the same gene. And thus, the full range of possibilities would be:
RR = normal black/brown/blond
Rr = normal black/brown/blond
Rs = normal black/brown/blond
ss = normal black/brown/blond
SS = silver
Sr = silver
Ss = silver
SR = ? If silver overrides normal pigmentation, I would suspect that would be silver, as well. But perhaps darker?
rs = ? red (but possibly a different shade....)
rr = red

Since we don't seem to have silver and red running in the same families, this shouldn't make any significant difference, but it would limit outcomes if that were to happen. What would be more limiting would be if there were not actually 4 different alleles. But since silver is being treated as simple dominant and red as simple recessive (and neither hair color is common outside certain families), it makes sense that the majority of the population would have some combination that gives 'normal' hair. So, it is not necessary to combine the traits onto a single gene like this. Just interesting to note that doing so would hardly disrupt anything! (And might allow the red and silver traits to 'interact' a bit...)
 
Last edited:

Faelivrin

Well-Known Member
Oi ve, my brain starts to lose it at all that. I've been treating silver as a simple recessive so far, but it could also be ... the things you described. :confused: But let's see...

Cirdan is white (has no close relatives, could be any #)
Olwe is white, could plausibly be any # from 0-9
Earwen is "star-like silver", could be 0-4
Miriel is silver, could be 4-10
Thingol is "grey silver", could be 5-10

In my current model silver won't affect hair of more than 6 eumelanin genes, so none of the above can have more than 6.
So if "silver" includes 4, white is some amount less than 4, and "grey silver" is some amount more than 4, there can be degrees of dilution in silver hair.
I would say 0-3 gives white, 4-5 gives silver, and 6-10 gives "grey silver".

EDIT: So I conclude Olwe is 0 to 3, Earwen is 4, Thingol is 6, Miriel is 5.
Making Finarfin and Earwen both 4's makes the chances of dark-haired kids 36.3%, though.


I did get the cross %s finished for my 5-gene model. Again, these %s assume maximum heterozygosity for everybody.

This forum can't process Google Drive urls. :( So here are the tables: https://drive.google.com/drive/folders/1qdGVc2wnnXGpwSLFwVtdFJKqu_4wcCik
Sorry that isn't easy to read without downloading them.

The chances of Nerdanel and Feanor having 2 or 3/4 kids with 5 is pretty low, so I'd make Celegorm 5 and Maedhros 6.
 
Last edited:

MithLuin

Administrator
Staff member
To be fair, the chances of Nerdanel and Feanor both being Rr for red and yet having 3/7 of their children be red-headed is also surprising. That result is more likely if Nerdanel is rr. And yet...she's not. It is still a 25 percent chance for each kid to get the red hair with her being a carrier.

I am fine assigning Nerdanel a 5 and Feanor a 9, leaving plenty of opportunity for a kid to take on the 5-6 shades. And if one kid does, it's equally likely that another could. Just because it's low odds doesn't mean they can only have one child like that.

I can probably make silver work as a recessive, as long as it's common enough among the Teleri to assume that everyone marrying into Elwe's family is at least a carrier. But I am more comfortable with the simple dominant inheritance pattern here. The only suspicious family so far is Finarfin's kids all being non-silver when they all had a 50/50 shot at it. Not that you can't flip heads 4 times in a row, but...
 
Last edited:

Faelivrin

Well-Known Member
But identical twins basically only count as one person, genetically speaking. So they got 2 redheads/6 zygotes. (The twins are epigenetically different but I won't try to figure out why.)

You can look at the chances of two 5's as 3.125%, then 3.125%, but overall the chance or both is 3.125 x 3.125 = 0.09766%

But 3% can happen at some point. That's basically the chances I give Turin to be a 9, but I'm leaning to make him one. (I'm working out Earendil's entire extended family right now. It goes so much faster now that I can just copy+paste %s from the giant table! :))
 
Last edited:

MithLuin

Administrator
Staff member
If Nerdanel has a pigmentation of 5 and Fëanor has a pigmentation of 9, then according to your chart above, the chances of them having a child with a darkness of 5 is actually 9%. That's one reason why I think Nerdanel should have such a light pigmentation. If she and Fëanor had more uniformly dark-haired children, I would assume that she has darker pigment. But with about half of the kids being quite light....I'm assigning her a 5 in the absence of any dilution factors.

I agree that it's unlikely that they'd get two children with a trait that is only a 3% chance, but keep in mind that the probability here is not additive. It's the probability for each child. If there's a 25% chance of a trait appearing, it's not more likely to appear in the 4th child than the 1st, even if the first three children do not show the trait. IF you flip heads 3 times in a row, you still have a 50/50 shot of flipping heads on the 4th try. Even if the chances of flipping heads 4 times in a row is 1/(2^4) = 1/16 = 6%.


I can make 0-2 white and 3-6 silver, with 7-10 being 'dilute'. That would allow Earwen to be a 3, while her father is a 2. I agree that 'starlight silver' sounds pretty pale, not greyish at all. So maybe white really is the palest shades only (0-1). I worried about making white = 0 because of how that constrains Thingol (as Olwë's brother), but once I figure out what we can get away with for him, I might be more comfortable with it. I see from your charts that it would not be difficult to make Olwë a 0 and still have Thingol be a 3 or even a 4.
 
Last edited:

Faelivrin

Well-Known Member
OK, so far in Earendil's family, I think:

Fingolfin = 8 dark brown
Anairë = 5-10 dark
Aredhel = 8-10 dark
Eöl = 8-10 dark
Maeglin = 10 deep black
Eärendil = 3-4 gold
Morwen = 6 dark (brown)
Húrin = 4 ash blond
Túrin = 9 black
Lalaeth
= 2-4 blond
Nienor = 2-4 blond
Galdor = 0-1 light blond
Hareth
= 8 dark brown


If Nerdanel has a pigmentation of 5 and Fëanor has a pigmentation of 9, then according to your chart above, the chances of them having a child with a darkness of 5 is actually 9%. That's one reason why I think Nerdanel should have such a light pigmentation. If she and Fëanor had more uniformly dark-haired children, I would assume that she has darker pigment. But with about half of the kids being quite light....I'm assigning her a 5 in the absence of any dilution factors
If Nerdanel is 5, yes. I was thinking 6 though.

I agree that it's unlikely that they'd get two children with a trait that is only a 3% chance, but keep in mind that the probability here is not additive. It's the probability for each child. If there's a 25% chance of a trait appearing, it's not more likely to appear in the 4th child than the 1st, even if the first three children do not show the trait. IF you flip heads 3 times in a row, you still have a 50/50 shot of flipping heads on the 4th try. Even if the chances of flipping heads 4 times in a row is 1/(2^4) = 1/16 = 6%.
But the event that two independent events both happen is P(1) x P(2).


I can make 0-2 white and 3-6 silver, with 7-10 being 'dilute'. That would allow Earwen to be a 3, while her father is a 2. I agree that 'starlight silver' sounds pretty pale, not greyish at all.
With 3 being gold and 4 ash-blond, I don't think diluting a 3 can produce anything even pale greyish. Unless you think "star-like silver" can be stark white.
 

MithLuin

Administrator
Staff member
Well, when you have 7 children, the chances of having whatever family you wind up with is astronomically smaller than the chances of any one child getting any one trait. Punnett squares (the percentages you calculated) generate the possibility of any given outcome for each time the cross is done (as you know). The only reason to pay attention to the combination of children in a family is if all of the kids are getting an unlikely combination, and none of the kids are getting the really common combination - because then the model may be wrong. But unless you're having thousands of kids, that's probably not statistically significant.

As a for instance, all 5 children in my family have blue eyes. While this would be completely understandable if both of my parents had blue eyes, they do not. My Dad has hazel eyes. So, there was always a chance that each kid could have not-blue eyes. And, multiply those chances by 5, and....the chances of all of us having blue eyes is quite small. But, we do. So, there's that.

My Dad has O- blood and my mom has A+. I'm A-. Knowing this, I can (almost) determine their genotypes.
Dad: oo Rh-Rh-
Mom: A* Rh+Rh-
IF all five of their kids had A blood, that would not prove that my mom is homozygous for A (AA). She could still be Ao, and just happened to pass on the A to every single one of her offspring. Having a child with oo blood would prove that she's a carrier for it, but failing to have a kid with that trait does not prove that she's not a carrier.

It's unlikely that a roulette wheel will land on black 10 times in a row. But it happens every day in the casinos.



I would say that's all looking really good, except for Morwen being only a 6. That seems rather light for her.
 
Last edited:

Ange1e4e5

Well-Known Member
Finarfin's kids aren't very constrained... Angrod needs to be light enough to have a blond kid and at least one blond grandkid. Galadriel has dark grandkids, but I'm increasingly inclined to credit Melian with simple dominant super-black hair.

Agreed Fingolfin is probably 8, but wait till I get to his grandchildren (unless you already have).

Still working on Turgon x Elenwe and planning to do Aredhel x Eol, but I'm now tired of calculating all these probabilities and want to program giant Punnett squares into Excel before I continue. This may take a while.
Well, what hair color would Eol have?
 

Faelivrin

Well-Known Member
I assumed Eol would be dark-haired like most Sindar, but "Dark Elf" seems to refer mainly to his hatred of the Sun and Moon.

Morwen needs to be a light-ish brown because she had 2 blond kids. Her mother was blond, too. I'm not sure I remember why I made her specifically 6, though - she could be a 7 or even 8, although 8 is pushing it. (Maybe I picked 6 based on faulty #s.)

Nerdanel = 5 and Morwen = 7 both make sense to me.
 

Ange1e4e5

Well-Known Member
I assumed Eol would be dark-haired like most Sindar, but "Dark Elf" seems to refer mainly to his hatred of the Sun and Moon.

Morwen needs to be a light-ish brown because she had 2 blond kids. Her mother was blond, too. I'm not sure I remember why I made her specifically 6, though - she could be a 7 or even 8, although 8 is pushing it.

Let me recalculate with Nerdanel = 5 and Morwen = 7.
I thought that Sindar were silver-haired, as part of the Teleri.
 
Top